Research importance

IN THE LAST DECADE OF DEVELOPMENT AND APPLICATION OF QUANTUM MECHANICAL METHODS as called great attention in science due to their capacity to explain physical and chemical phenomena at the microscopic level with a high degree of accuracy.

Density Functional Theory is one of the latest methods characterized as a universal approach and used to study different types of systems such as atoms, molecules and extended systems (i.e. polymers, slabs and crystals). This theory (DFT) can predict a variety of structural and electronic properties which are of fundamental importance for the proper understanding of matter behavior. For instance, we can get information about:

• Molecular geometries (bond distances, bond angles, rotational constants)
• Energies (interaction, atomization, orbital, ionization, affinity, additive, nonadditive, zero-point, reorganization, activation, band gaps, etc.)
• Volumetric properties (spin polarized density and spin compensated density), electrostatic potential, molecular orbitals)
• Charges (net atomic populations, bond orders)
• Multipole moments (dipole, quadrupole, octupole, etc.)
• Vibration spectra (frequencies, intensities and normal vibration modes)

DFT is formulated on relatively simple theoretical grounds and, from a computational perspective, it is one of the most flexible methods to apply. However, the present predictions of the theory are found in the limits of chemical accuracy, with no clear way to systematically improve it. Moreover, DFT is still limited to a few dozens of heavy atoms or a few hundreds of light atoms.

Contribution to the research

OUR INTEREST RESIDES ON INCREASING THE ACCURACY OF THE DFT SCHEME and make its applications feasible to compounds of larger dimensions. In this regard, we were able to design the first density functional expression for the exchange energy which incorporates a truely nonlocal operator of the 1/r12 type. It was shown that expressions like this are capable to yield energies of comparable accuracy to those functionals that make use of electron density gradients and which have been wrongly called nonlocal type expressions. In a recent review (D. G. Truhlar, First International Conference on Foundations of Molecular Modeling and Simulation, Technical Program & Preprints, Keystone Resort, Co. USA, July 23-28, 2000) it has been pointed out the necessity to incorporate the 1/r12 nonlocality.

Recently, we have developed a new method, VEDA (acronym of Virtual Electron Density Approach) with the goal of investigating large compounds mainly of biological interest. The method is based on the fragmentation of a molecule and the construction of an electron density from fragments that virtually does the work of an SCF electron density. VEDA can be used in the frame of the KS-DFT theory and does not require reconstructing every atomic orbital from scratch. The method takes advantage of smaller calculations on fragments to propose a new wavefunction and proceed to the evaluation of the energy as usual. In this context molecular partitions are important and can take part of a data bank for later use in the calculations, allowing the method to become more efficient. In VEDA all atoms are treated at the quantum level and it can be combined with other schemes like molecular mechanics for the use of hybrid approaches. VEDA loses some accuracy in the fragmentation process, specially at fragment joints, but this loss is small (less than 2 kcal) and more importantly, it is under control, so it can be further reduced. To our best knowledge this is the first time that DFT energies can be obtained without having recourse to the SCF process.

Benefits for the Scientific Community and Society

IN TRADITIONAL AB-INITIO METHODS THE EXCHANGE AND CORRELATION ENERGIES re the main obstacles to apply quantum theory to large systems. Density functional theory is able to overcome such obstacles by the use of alternative expressions for exchange and correlation. If we are able to suggest improvements to such expressions in the DFT frame, then it is possible to achieve more accurate structural and electronic features of a variety of systems and processes of interest. Also, the development of new approaches like VEDA opens new application fields of quantum physics to investigate large compounds in their molecular environment.

The use of more realistic computational models let us close gaps between theory and experiment. By implementing VEDA we have tried to preserve the fundamentals of DFT, where the electron density is supposed to play the central role in defining the molecular properties. Finally, the building of a solid, reliable and computationally efficient theory leads us to a better understanding of the problems that have a molecular origen and which usually manifest at the macroscopic level. Currently, important problems are under investigation around the world in areas like molecular genetics (ribozymes, cancer, HIV), solid-state physics (nanoelectronic circuits), the chemical industry (new catalytic materials), etc. Such investigations are of paramount interest for our society.

Future projection

GIVEN THE SCIENTIFIC IMPORTANCE OF THE RESEARCH, our work is to follow the same line but with new objectives. It is our aim to design an accurate and flexible correlation density functional expression that can be easily implemented and computed. In parallel form, we will improve VEDA to make it faster. Our ultimate goal is to apply quantum physics to large sytems with a minimum lost of accuracy, at a reasonable computational cost.